EXECUTIVE SUMMARYClipper Windpower, in collaboration with United Technologies Research Center, the National Renewable Energy Laboratory, and Hamilton Sundstrand Corporation, developed a low-cost, deflection-compliant, reliable, and serviceable chain drive speed increaser.This chain and sprocket drivetrain design offers significant breakthroughs in the areas of cost and serviceability and addresses the key challenges of current geared and direct-drive systems. The use of gearboxes has proven to be challenging; the large torques and bending loads associated with use in large multi-MW wind applications have generally limited demonstrated lifetime to 8-10 years [1]. The large cost of gearbox replacement and the required use of large, expensive cranes can result in gearbox replacement costs on the order of $1M, representing a significant impact to overall cost of energy (COE). Direct-drive machines eliminate the gearbox, thereby targeting increased reliability and reduced life-cycle cost. However, the slow rotational speeds require very large and costly generators, which also typically have an undesirable dependence on expensive rare-earth magnet materials and large structural penalties for precise air gap control. The cost of rare-earth materials has increased 20X in the last 8 years representing a key risk to ever realizing the promised cost of energy reductions from direct-drive generators. A common challenge to both geared and direct drive architectures is a limited ability to manage input shaft deflections.The proposed Clipper drivetrain is deflection-compliant, insulating later drivetrain stages and generators from off-axis loads. The system is modular, allowing for all key parts to be removed and replaced without the use of a high capacity crane. Finally, the technology modularity allows for scalability and many possible drivetrain topologies. These benefits enable reductions in drivetrain capital cost by 10.0%, levelized replacement and O&M costs by 26.7%, and overall cost of energy by 10.2%.This design was achieved by: (1) performing an extensive optimization study that determined the preliminary cost for all practical chain drive topologies to ensure the most competitive configuration; (2) conducting detailed analysis of chain dynamics, contact stresses, and wear and efficiency characteristics over the chain's life to ensure accurate physics-based predictions of chain performance; and (3) developing a final product design, including reliability analysis, chain replacement procedures, and bearing and sprocket analysis. Definition of this final product configuration was used to develop refined cost of energy estimates. Finally, key system risks for the chain drive were defined and a comprehensive risk reduction plan was created for execution in Phase 2.